Multi-drug combinations for the treatment of complex diseases are gradually becoming an important treatment, and this type of treatment can take advantage of the synergistic effects among drugs. However, drug-drug interactions (DDIs) are not just all beneficial. Accurate and rapid identifications of the DDIs are essential to enhance the effectiveness of combination therapy and avoid unintended side effects. Traditional DDIs prediction methods use only drug sequence information or drug graph information, which ignores information about the position of atoms and edges in the spatial structure. In this paper, we propose Molormer, a method based on a lightweight attention mechanism for DDIs prediction. Molormer takes the two-dimension (2D) structures of drugs as input and encodes the molecular graph with spatial information. Besides, Molormer uses lightweight-based attention mechanism and self-attention distilling to process spatially the encoded molecular graph, which not only retains the multi-headed attention mechanism but also reduces the computational and storage costs. Finally, we use the Siamese network architecture to serve as the architecture of Molormer, which can make full use of the limited data to train the model for better performance and also limit the differences to some extent between networks dealing with drug features. Experiments show that our proposed method outperforms state-of-the-art methods in Accuracy, Precision, Recall and F1 on multi-label DDIs dataset. In the case study section, we used Molormer to make predictions of new interactions for the drugs Aliskiren, Selexipag and Vorapaxar and validated parts of the predictions. Code and models are available at https://github.com/IsXudongZhang/Molormer.

Molormer: a lightweight self-attention-based method focused on spatial structure of molecular graph for drug-drug interactions prediction

Protein phosphorylation is one of the most critical post-translational modifications of proteins in eukaryotes, which is essential for a variety of biological processes. Plenty of attempts have been made to improve the performance of computational predictors for phosphorylation site prediction. However, most of them are based on extra domain knowledge or feature selection. In this article, we present a novel deep learning-based predictor, named TransPhos, which is constructed using a transformer encoder and densely connected convolutional neural network blocks, for predicting phosphorylation sites. Data experiments are conducted on the datasets of PPA (version 3.0) and Phospho. ELM. The experimental results show that our TransPhos performs better than several deep learning models, including Convolutional Neural Networks (CNN), Long-term and short-term memory networks (LSTM), Recurrent neural networks (RNN) and Fully connected neural networks (FCNN), and some state-of-the-art deep learning-based prediction tools, including GPS2.1, NetPhos, PPRED, Musite, PhosphoSVM, SKIPHOS, and DeepPhos. Our model achieves a good performance on the training datasets of Serine (S), Threonine (T), and Tyrosine (Y), with AUC values of 0.8579, 0.8335, and 0.6953 using 10-fold cross-validation tests, respectively, and demonstrates that the presented TransPhos tool considerably outperforms competing predictors in general protein phosphorylation site prediction.

/pdf/transphos-a-deep-learning-model-for-general-phosphorylation-231cyiom.pdf

TransPhos: A Deep-Learning Model for General Phosphorylation Site Prediction Based on Transformer-Encoder Architecture

A mixed reality-based navigation method for dental implant navigation method: A pilot study

RNA plays an important role in biology, and more than 170 RNA modifications have been identified so far. Post-transcriptional modification of RNA in cells plays a crucial role in the regulation of its stability, transport, processing, and gene expression. So far, the research on RNA modification and the exact role of its enzymes is becoming more and more comprehensive. Human immunodeficiency virus 1 (HIV-1) is an RNA virus and the causative agent of acquired immunodeficiency syndrome (AIDS), which is one of the most devastating viral pandemics in history. More and more studies have shown that HIV has RNA modifications and regulation of its gene expression during infection and replication. This review focuses on several RNA modifications and their regulatory roles as well as the roles that different RNA modifications play during HIV-1 infection, in order to find new approaches for the development of anti-HIV-1 therapeutics.

/pdf/the-role-of-rna-modification-in-hiv-1-infection-djzahwia.pdf

The Role of RNA Modification in HIV-1 Infection

Objective
AlkB homolog 5 (ALKBH5) has been proven to be closely related to tumors. However, the role and molecular mechanism of ALKBH5 in neuroblastomas have rarely been reported.


Methods
The potential functional single-nucleotide polymorphisms (SNPs) in ALKBH5 were identified by National Center for Biotechnology Information (NCBI) dbSNP screening and SNPinfo software. TaqMan probes were used for genotyping. A multiple logistic regression model was used to evaluate the effects of different SNP loci on the risk of neuroblastoma. The expression of ALKBH5 in neuroblastoma was evaluated by Western blotting and immunohistochemistry (IHC). Cell counting kit-8 (CCK-8), plate colony formation and 5-ethynyl-2'-deoxyuridine (EdU) incorporation assays were used to evaluate cell proliferation. Wound healing and Transwell assays were used to compare cell migration and invasion. Thermodynamic modelling was performed to predict the ability of miRNAs to bind to ALKBH5 with the rs8400 G/A polymorphism. RNA sequencing, N6-methyladenosine (m6A) sequencing, m6A methylated RNA immunoprecipitation (MeRIP) and a luciferase assay were used to identify the targeting effect of ALKBH5 on SPP1.


Results
ALKBH5 was highly expressed in neuroblastoma. Knocking down ALKBH5 inhibited the proliferation, migration and invasion of cancer cells. miR-186-3p negatively regulates the expression of ALKBH5, and this ability is affected by the rs8400 polymorphism. When the G nucleotide was mutated to A, the ability of miR-186-3p to bind to the 3'-UTR of ALKBH5 decreased, resulting in upregulation of ALKBH5. SPP1 is the downstream target gene of the ALKBH5 oncogene. Knocking down SPP1 partially restored the inhibitory effect of ALKBH5 downregulation on neuroblastoma. Downregulation of ALKBH5 can improve the therapeutic efficacy of carboplatin and etoposide in neuroblastoma.


Conclusions
We first found that the rs8400 G>A polymorphism in the m6A demethylase-encoding gene ALKBH5 increases neuroblastoma susceptibility and determines the related mechanisms. The aberrant regulation of ALKBH5 by miR-186-3p caused by this genetic variation in ALKBH5 promotes the occurrence and development of neuroblastoma through the ALKBH5-SPP1 axis.

Variant rs8400 enhances ALKBH5 expression through disrupting miR-186 binding and promotes neuroblastoma progression.

N6-methyladenosine (m6A) RNA modification plays a critical role in various physiological and pathological conditions. However, the role of m6A modification in head and neck squamous cell carcinoma (HNSCC) remains elusive.In this study, the expression of m6A demethylases was detected by HNSCC tissue microarray. m6A-RNA immunoprecipitation (MeRIP) sequencing and RNA sequencing were used to identify downstream targets of ALKBH5. Comprehensive identification of RNA-binding proteins by mass spectrometry (ChIRP-MS) was used to explore the m6A "readers". Tumor-infiltrating lymphocytes were analyzed in SCC7-bearing xenografts in C3H mice.Here, we demonstrate the downregulation of m6A status and upregulation of two demethylases in HNSCC. Silencing the m6A demethylase alkB homolog 5, RNA demethylase (ALKBH5) suppresses tumor progression in vitro and in vivo. m6A-RNA immunoprecipitation sequencing reveals that ALKBH5 downregulates the m6A modification of DDX58 mRNA. Moreover, RIG-I, encoded by the DDX58 mRNA, reverses the protumorigenic characteristics of ALKBH5. ChIRP-MS demonstrates that HNRNPC binds to the m6A sites of DDX58 mRNA to promote its maturation. ALKBH5 overexpression inhibits RIG-I-mediated IFNα secretion through the IKKε/TBK1/IRF3 pathway. The number of tumor-infiltrating lymphocytes in C3H immunocompetent mice is reduced by ALKBH5 overexpression and restored by IFNα administration. Upregulation of AKLBH5 negatively correlates with RIG-I and IFNα expression in HNSCC patients.These findings unveil a novel mechanism of immune microenvironment regulation mediated by m6A modification through the ALKBH5/RIG-I/IFNα axis, providing a rationale for therapeutically targeting epitranscriptomic modulators in HNSCC. 

/pdf/the-m6a-demethylase-alkbh5-promotes-tumor-progression-by-1ctws3d5.pdf

The m6A demethylase ALKBH5 promotes tumor progression by inhibiting RIG-I expression and interferon alpha production through the IKKε/TBK1/IRF3 pathway in head and neck squamous cell carcinoma

There is a strong need to eliminate batch-specific differences when integrating single-cell RNA-sequencing (scRNA-seq) datasets generated under different experimental conditions for downstream task analysis. Existing batch correction methods usually transform different batches of cells into one preselected “anchor” batch or a low-dimensional embedding space, and cannot take full advantage of useful information from multiple sources. We present a novel framework, called IMGG, i.e., integrating multiple single-cell datasets through connected graphs and generative adversarial networks (GAN) to eliminate nonbiological differences between different batches. Compared with current methods, IMGG shows excellent performance on a variety of evaluation metrics, and the IMGG-corrected gene expression data incorporate features from multiple batches, allowing for downstream tasks such as differential gene expression analysis.

https://www.mdpi.com/1422-0067/23/4/2082/pdf?version=1644831392

IMGG: Integrating Multiple Single-Cell Datasets through Connected Graphs and Generative Adversarial Networks

Cancer prognosis is an essential goal for early diagnosis, biomarker selection, and medical therapy. In the past decade, deep learning has successfully solved a variety of biomedical problems. However, due to the high dimensional limitation of human cancer transcriptome data and the small number of training samples, there is still no mature deep learning-based survival analysis model that can completely solve problems in the training process like overfitting and accurate prognosis. Given these problems, we introduced a novel framework called SAVAE-Cox for survival analysis of high-dimensional transcriptome data. This model adopts a novel attention mechanism and takes full advantage of the adversarial transfer learning strategy. We trained the model on 16 types of TCGA cancer RNA-seq data sets. Experiments show that our module outperformed state-of-the-art survival analysis models such as the Cox proportional hazard model (Cox-ph), Cox-lasso, Cox-ridge, Cox-nnet, and VAECox on the concordance index. In addition, we carry out some feature analysis experiments. Based on the experimental results, we concluded that our model is helpful for revealing cancer-related genes and biological functions.

/pdf/a-novel-attention-mechanism-based-cox-survival-model-by-35nine44.pdf

A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information

5,7-Dihydroxy-4-methylcoumarin (D4M) is attributed to free radical scavenging effects, with wide application for anti-oxidation. This work aimed to assess D4M's impact on cisplatin-induced ototoxicity. The cell viability was estimated with CCK-8 assay. Apoptosis was detected by the Annexin V-FITC and PI assay. The reactive oxygen species (ROS) level was determined by MitoSOX-Red and CellROX-Green probes. Mitochondrial membrane potential was analyzed with TMRM staining. Immunofluorescence was utilized for hair cells and spiral ganglion neuron detection. Apoptosis-associated proteins were assessed by cleaved caspase-3 and TUNEL staining. These results showed that D4M pretreatment protected hair cells from cisplatin-induced damage, increased cell viability, and decreased apoptosis in House Ear Institute-Organ of Corti1 (HEI-OC1) cells and neonatal mouse cochlear explants. D4M significantly inhibited cisplatin-induced mitochondrial apoptosis and reduced ROS accumulation. In addition, the protective effect of D4M on cisplatin-induced ototoxicity was also confirmed in cochlear hair cells and spiral ganglion neurons in neonatal mice. Mechanistic studies showed that D4M markedly downregulated p-JNK and elevated the expression ratio of p-FoxO1/FoxO1, thereby reducing cisplatin-induced caspase-dependent apoptosis. Meanwhile, D4M-related protection of HEI-OC1 cells was significantly blunted by JNK signaling induction with anisomycin. This study supports the possibility that D4M may be used as a new compound to prevent cisplatin-related hearing loss. 

5,7-Dihydroxy-4-methylcoumarin modulates the JNK/FoxO1 signaling pathway to attenuate cisplatin-induced ototoxicity by suppressing oxidative stress and apoptosis in vitro.

Zhiyuan Zhang

Papers

The m6A demethylase ALKBH5 promotes tumor progression by inhibiting RIG-I expression and interferon alpha production through the IKKε/TBK1/IRF3 pathway in head and neck squamous cell carcinoma

TransPhos: A Deep-Learning Model for General Phosphorylation Site Prediction Based on Transformer-Encoder Architecture

IMGG: Integrating Multiple Single-Cell Datasets through Connected Graphs and Generative Adversarial Networks

A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information

5,7-Dihydroxy-4-methylcoumarin modulates the JNK/FoxO1 signaling pathway to attenuate cisplatin-induced ototoxicity by suppressing oxidative stress and apoptosis in vitro.